26417 4th pages / 1 of 13

ABSTRACT

The West Antarctic Rift System provides critical geological boundary conditions for the overlying West Antarctic Ice Sheet. Pre-vious geophysical surveys have traced the West Antarctic Rift System and addressed the controls that it exerts on the West Ant-arctic Ice Sheet in the Ross Sea Embayment. However, much less is known about the rift system under the Amundsen Sea Embay-ment, a key sector of the West Antarctic Ice Sheet, which is thinning signifi cantly today. New aero gravity data over the Pine Island Glacier region, one of the fastest fl owing gla-ciers within the Amundsen Sea Embayment, sheds new light into the crustal structure under this dynamic part of the West Ant-arctic Ice Sheet. Three-dimensional (3-D) in-version of terrain-decorrelated free-air and Bouguer gravity anomaly data reveal signifi -cant crustal thinning beneath the catchment of Pine Island Glacier. Under the Byrd Sub-glacial Basin and the newly identifi ed Pine Island Rift, Moho depth is estimated to be 19 ± 1 km. This is the thinnest crust observed beneath the West Antarctic Ice Sheet. Esti-mates of lithosphere rigidity (Te), based on isostatic models, yield a Te of 5 ± 5 km, which is comparable to values from modern rift sys-tems such as the Basin and Range Province. Major crustal thinning, coupled with low lithosphere rigidity, attest to the considerable impact of continental rifting beneath this part of the West Antarctic Ice Sheet. In analogy with the better known Ross Sea segment of the West Antarctic Rift System we suggest that the Amundsen Sea Embayment was af-fected by distributed Cretaceous rifting, fol-lowed by Cenozoic narrow-mode rifting. Narrow-mode rifting within the Pine Island Rift is particularly important as it may serve as a geological template for enhanced glacial fl ow associated with Pine Island Glacier.

INTRODUCTION

The West Antarctic Rift System (Fig. 1) is thought to extend from the Ross Sea to the Bellingshausen Sea and has been compared to other major continental rift zones, such as the Basin and Range Province or the East African Rift System (LeMasurier and Thomson, 1990; LeMasurier and Landis, 1996; Winberry and Anandakrishnan, 2003; LeMasurier, 2008). While the Ross Sea segment of the West Ant-arctic Rift System is relatively well understood, due to extensive seismic and potential-fi eld exploration (Damaske et al., 1994; Brancolini et al., 1995; Behrendt, 1999) coupled with drilling (Cape-Roberts-Science-Team, 2000; Fielding et al., 2006), a greater degree of uncer-tainty surrounds the reminder of the rift system. The reasons for this are that it is obscured by the West Antarctic Ice Sheet and that current geophysical surveys do not provide compre-hensive coverage (Dalziel, 2006).

The location of the West Antarctic Rift Sys-tem beneath the West Antarctic Ice Sheet is highly signifi cant because bedrock geology, and the crustal structure of the rift system, may in-fl uence the dynamics of the overlying glaciers, which in turn affect the stability of the ice sheet (Dalziel and Lawver, 2001; Rignot et al., 2008). The suggestion that the rift system may provide critical boundary conditions on ice dynamics is supported by geophysical studies over the Ross Sea Embayment sector of the West Antarctic Rift System. For example, airborne geophysics re-vealed a rift basin with sedimentary infi ll under the onset region of Ice Stream B (Bell et al., 1998). Airborne radar and aeromagnetic investi-gations have led to the suggestion that active or recently active subglacial volcanism associated with the West Antarctic Rift System may also promote enhanced glacial fl ow by increasing the availability of water beneath the Ross Sea Em-bayment region (Blankenship et al., 1993; Hulbe and MacAyeal, 1999; Vogel and Tulaczyk, 2006).

Recent studies show that parts of the West Antarctic Ice Sheet are thinning, with the most

dramatic changes occurring in the Amundsen Sea Embayment (Davis et al., 2005). Pine Island Glacier is one of the fastest fl owing glaciers in the Amundsen Sea Embayment (Fig. 2) and is thinning and retreating rapidly, contributing to accelerated global sea-level rise (Rignot, 1998; Shepherd et al., 2004; Thomas et al., 2004). However, the geological boundary condi-tions for this highly dynamic part of the West Antarctic Ice Sheet remain poorly constrained (Ferraccioli et al., 2007), despite the availabil-ity of subice information from recent radio echo sounding and gravity data (Holt et al., 2006a; Vaughan et al., 2006; Diehl et al., 2008).

Here we focus on utilizing newly acquired airborne gravity data to model the crustal struc-ture of the West Antarctic Rift System beneath the catchment of Pine Island Glacier (Fig. 2). A variety of 3-D inversion techniques provide new estimates of crustal thickness under this poorly known sector of the West Antarctic Rift System. In addition, we compare the observed grav-ity data with local and regional isostatic com-pensation models to derive the rigidity of the rifted lithosphere (Te). Our crustal models for the Amundsen Sea Embayment region reveal the occurrence of multiple continental rifting stages, including wide-mode Cretaceous rifting, followed by narrow-mode Cenozoic rifting. We suggest that the latter rifting stage may be of signifi cant relevance for enhanced glacial fl ow associated with Pine Island Glacier.

GEOLOGICAL AND GEOPHYSICAL SETTING

Cretaceous rifting within the West Antarctic Rift System is thought to have been caused by ridge-trench collision and progressive west-erly cessation of subduction along the paleo-Pacifi c margin of Gondwana between 110 and 94 Ma (Mukasa and Dalziel, 2000; Larter et al., 2002). This passive rifting event may have led to the eventual separation of New Zealand and Chatham Rise from Antarctica ca. 85 Ma (Luyendyk , 1995). Alternatively, the early rift

For permission to copy, contact editing@geosociety.org© 2009 Geological Society of America

1

GSA Bulletin; Month/Month 2009; v. 1xx; no. X/X; p. 1–13; doi: 10.1130/B26417.1; 8 fi gures; 1 table.

†E-mail: tomj@bas.ac.uk

Aerogravity evidence for major crustal thinning under the Pine Island Glacier region (West Antarctica)

T.A. Jordan1,†, F. Ferraccioli1, D.G. Vaughan1, J.W. Holt2, H. Corr1, D.D. Blankenship2, and T.M. Diehl2

1British Antarctic Survey, High Cross, Madingley Road, Cambridge, CB3 0ET, UK

2University of Texas, Institute of Geophysics, 10100 Burnet Road, Austin, Texas 78758-4445, USA

Jordan et al.

2 Geological Society of America Bulletin, Month/Month 2009

26417 4th pages / 2 of 13

stage may have been actively driven by a Cre-taceous mantle plume centered beneath Marie Byrd Land (Weaver et al., 1994).

Further evidence for Cretaceous intraconti-nental extension is seen in mylonitic gneisses dredged from the eastern Ross Sea Rift, which yield deformation ages of 98–95 Ma. These

rocks also yield U-Pb zircon emplacement ages of 102–97 Ma, which can be correlated onshore to the A-type Byrd Coast Granite, which is re-lated to continental extension (Siddoway et al., 2004). On the Marie Byrd Land coast mafi c magmatism occurred at ca. 107 Ma and may be linked to the presence of a mantle plume (Storey

et al., 1999). Structural evidence in the Ford Ranges suggests Cretaceous transtensional rift-ing (Siddoway et al., 2005), and aeromagnetic patterns indicate subglacial Cretaceous mafi c magmatism and block faulting on the Edward VII Peninsula (Ferraccioli et al., 2002). Fur-ther to the east, offshore from Pine Island Gla-cier, two geophysical features are identifi ed, thought to be associated with ca. 90 Ma rifting and later deformation along the margin of the Bellingshausen Plate between 79 and 61 Ma (Fig. 1) (Gohl et al., 2007).

After the initial Cretaceous phases of rifting there appears to have been a period of tectonic quiescence in Marie Byrd Land leading to the development of a low-relief erosion surface around 85–75 Ma (LeMasurier and Landis, 1996). Rift-related alkaline magmatism began at ca. 50 Ma in Northern Victoria Land (Fig. 1) (Rocchi et al., 2002). However, in Marie Byrd Land the oldest Cenozoic intrusive magma-tism is ca. 34 Ma (Rocchi et al., 2006), and the main phase of volcanism began at 28–30 Ma (LeMasurier and Thomson, 1990). Pliocene and recently active volcanism in Marie Byrd Land and the Ross Sea region has continued to the present-day (LeMasurier and Thomson, 1990). In the Hudson Mountains, on the southeastern edge of the Thurston Island block (Fig. 1), out-crops of Late Miocene and Pliocene subaerial basaltic lavas and hyaloclastites are also ob-served (LeMasurier and Thomson, 1990). There is also evidence of recent (ca. 2 ka) volcanism in the Hudson Mountains, based on interpreta-tion of a bright refl ector within the ice, identi-fi ed from new airborne radar data (Corr and Vaughan, 2008).

Evidence for extensive magmatism of in-ferred Cenozoic age beneath the West Antarctic Ice Sheet derives mainly from aeromagnetic data (Behrendt et al., 1996). This volcanism could be linked to oceanic rifting in the Adare Trough (Cande et al., 2000), which has been directly linked to continental rifting and magmatism within the Ross Sea Rift (Davey et al., 2006). Al-ternatively Cenozoic magmatism may relate to transtension in the Ross Sea Rift region (Ferrac-cioli et al., 2000; Rocchi et al., 2002). However, the widely distributed Cenozoic magmatism beneath the West Antarctic Ice Sheet is unlikely to result solely from far-fi eld rifting or transten-sional processes localized within the Ross Sea Rift. A Cenozoic mantle plume underlying some or all of the West Antarctic Rift System has been proposed by several authors (Behrendt et al., 1996; Behrendt, 1999; Storey et al., 1999). The presence of a plume explains the geochemical signatures and the generation of such large vol-umes of Cenozoic magmatism and is not in dis-agreement with the modest intraplate Cenozoic

90°

AmundsenSea

WeddellSea

RossSeaRift

CV

EB

?

?

East AntarcticaAP

TI

MBL

EWM

SipleCoast

–60°

–30°0°

30°

BS

B

BSTBel

lings

haus

enSe

a

AT

TransantarcticM

oun

t ains

VLB

NVL

PIG

ThwaitesHM

West Antarctic

Rift System

90°

AmundsenSea

WeddellSea

RossSeaRift

CV

EB

?

?

East AntarcticaAP

TI

MBL

EWM

SipleCoast

–60°

–30°0°

30°

BS

B

BSTBel

lings

haus

enSe

a

AT

TransantarcticM

oun

t ains

VLB

NVL

PIG

ThwaitesHM

(km)250 0 250

–5000 –800 0 3000 m

West Antarctic

Rift System

Figure 1. Regional tectonic sketch map showing the crustal setting of West Antarctica and the Pine Island Glacier (PIG) aerogeophysical survey area, overlaid on bedrock ele-va tions derived from BEDMAP (Lythe et al., 2000). The outline of the PIG catchment is traced in black. Tectonic blocks from Dalziel and Elliot (1982) in dotted white, namely: AP—Antarctic Peninsula, EWM—Ellsworth-Whitmore Mountains, TI—Thurston Island block, MBL—Marie Byrd Land. Possible extensions of the West Antarctic Rift System into the Amundsen or the Bellingshausen Sea Embayments are thick white lines (Dalziel, 2006). Sedi mentary basins inferred within the West Antarctic Rift System from previous aerogeophysical surveys are shown in thin white and are taken from Bell et al. (2006). Tri-angles mark Cenozoic volcanoes (LeMasurier and Thomson, 1990). Black triangles show in active volcanoes and white triangles are recently active ones. Abbreviations as follows: EB—Mount Erebus; CV—Mount CASERTZ, subglacial volcano; HM—Hudson Moun-tains; Victoria Land Basin (VLB); NVL—Northern Victoria Land; AT—Adare Trough; BSB—Byrd Subglacial Basin; BST—Bentley Subglacial Trench. Black lines offshore PIG mark geophysical features associated with ca. 90 Ma rifting, and deformation along the margin of the Bellingshausen Plate 79–61 Ma (Gohl et al., 2007). Black hexagons—location of passive seismic stations (Winberry and Anandakrishnan, 2004); white hexagon is the only passive seismic station within our aerogeophysical survey area.

Aerogravity evidence for major crustal thinning in West Antarctica

Geological Society of America Bulletin, Month/Month 2009 3

26417 4th pages / 3 of 13

crustal extension in West Antarctica predicted by some plate-tectonic reconstructions (Lawver and Gahagan, 1994; Steinberger et al., 2004). However, an alternate hypothesis invokes the existence of a diffuse alkaline magmatic prov-ince linked to detachment in Eocene times of late Cretaceous subducted slabs (Finn et al., 2005).

Major debate centers on the relative amount of Cenozoic extension in West Antarctica. Marine geophysical evidence over the Adare Trough suggest ~180 km of extension between 43 and 26 Ma (Cande et al., 2000). There is little evidence for major subsequent structural activity in the Adare Trough indicating that this part of the West Antarctic Rift system became largely inactive in the early Miocene (Müller et al., 2005). Extension within the Adare Trough may link, via intracontinental transfer faulting, to continental rifting in the western Ross Sea Rift (Davey et al., 2006). Alternatively Ceno-zoic crustal extension may be more diffuse and may have also affected the Central and Eastern basins of the Ross Sea Rift (Decesari et al., 2007). Comparison between the subice topog-raphy of the West Antarctic Rift System and other major rift systems has led to the sugges-tion that Neogene rifting may have affected the Byrd Subglacial Basin and the Bentley Sub-glacial Trench (LeMasurier, 2008).

It is unclear if parts of West Antarctic Rift Sys-tem are currently active or not (Dalziel, 2006). A single magnetotelluric experiment within the Byrd Subglacial Basin (Fig. 1) suggests that, at least locally, the lower crust and upper mantle of the West Antarctic Rift System has much higher resistivity, compared to other modern rift sys-tems (Wannamaker et al., 1996). Low earthquake magnitude and localized distribution of seismic events beneath the West Antarctic Rift System, compared to other major active continental rifts, also suggests that there may be limited active rifting (Winberry and Anandakrishnan, 2003).

SURVEY DESIGN AND DATA REDUCTION

The British Antarctic Survey conducted an airborne geophysical survey during the 2004–2005 fi eld season over the Pine Island Glacier catchment area (Fig. 2). A relatively coarse line and tie line spacing (~30 km) was selected to en-able, together with the U.S.-led AGASEA aero-geophysical survey, reconnaissance mapping of subglacial topography, geology, and ice-sheet structure across the entire Amundsen Sea Em-bayment in a single fi eld season (Fig. 3A) (Holt et al., 2006a; Vaughan et al., 2006). Our survey produced ~30,000 line-km of new airborne

gravity data. In addition, a detailed draped air-borne radar and aeromagnetic survey was fl own, over the fast-fl owing trunk of Pine Island Gla-cier (Fig. 2). Airborne gravity data were not col-lected on this higher resolution survey because of the vertical accelerations associated with draped fl ying.

The regional gravity survey was fl own in four constant elevation blocks (Fig. 2). This survey design resulted in a minimum ground clearance, important for recovery of high-quality radar and magnetic data, without changing altitude within a block. Aircraft altitude changes, which pro-duce higher accelerations and make recording accurate gravity data diffi cult, were restricted to turns at the end of lines. The Twin Otter aircraft used for the survey generally fl ew at ~60 m/s along the fl ight lines and completed the survey in a series of ~1000-km sorties. An overlap of ~30 km of level fl ight was allowed across survey block boundaries to ensure continuous coverage of useable gravity data. Additionally overlap-ping lines allowed for removal of fi ltering arti-facts at the ends of lines (Bell et al., 1999).

Airborne gravity data were collected using LaCoste and Romberg air-sea gravimeter S-83 (LaCoste, 1967), modifi ed and updated by the Zero Length Spring Corporation. The meter was mounted in a gyro-stabilized, shock-mounted platform, close to the center of mass of the air-craft to minimize the detrimental effect of vibra-tions and rotational motions on the meter. The meter platform operated with a 4 min period, yielding an acceptable trade-off between damp-ing of horizontal accelerations and time required for the platform to stabilize (Valliant, 1992).

All the sorties started and ended at the same fi eld camp and tie-down location (77°34.2′ S, 95°55.7′ W). Absolute gravity readings at this location were independently tied to absolute base stations at Rothera Station (Jones and Ferris , 1999) and McMurdo Station (Sasagawa et al., 2004) using LaCoste and Romberg land gravity meters (Diehl et al., 2008). The tie from Rothera gave a value of 982,446.0373 mGal at the Pine Island fi eld camp, while the tie from McMurdo (via a fi eld camp on Thwaites Gla-cier) gave a value of 982,446.682 mGal. The rela tive mismatch between the ties is signifi -cantly less than the typical accuracies for an airborne gravity survey (Bell et al., 1999). The McMurdo tie was preferred as there was less time between observations at the Thwaites and Pine Island fi eld camps than between observa-tions at Rothera and the Pine Island camp.

Standard processing steps were taken to con-vert the raw airborne gravity data to free-air anomalies (Jones and Johnson, 1995; Jones et al., 2002; Ferraccioli et al., 2005; Ferraccioli et al., 2006; Jordan et al., 2007). This included

1900 m2400 m

1400 m

2200 m

HudsonMountains

0 50 10050(km)

1900 m2400 m

1400 m

2200 m

HudsonMountains

EllsworthMountains

95°W

105°W

76°S 78°S

90°W

100°W

0 20 80 200 2000m/a

N

Figure 2. Ice velocity derived from InSAR data (Rignot et al., 2004) across the catchment of the Pine Island Glacier (black outline). Thin black lines mark our aerogeophysical survey fl ight lines. Dashed black and white lines separate different survey blocks and the bold numbers indicate the respective fl ight elevations. Dark gray outlines mark rock outcrops.

Jordan et al.

4 Geological Society of America Bulletin, Month/Month 2009

26417 4th pages / 4 of 13

making latitude, free-air, vertical acceleration, and Eötvös corrections (Harlan, 1968; Woollard, 1979) based on differential, carrier-phase, kine-matic global positioning system (GPS) positional data (Mader, 1992). In addition to the standard processing steps, mentioned previously, a cor-rection was made for gravimeter reading errors caused by the platform tilting, when it was sub-jected to horizontal accelerations (Swain, 1996).

The raw free-air gravity anomaly remains dominated by short-wavelength noise even after removal of all dynamic corrections (Bell et al., 1999; Jordan et al., 2007). Filtering is therefore required to minimize the residual short-wavelength noise. Two fi lters were con-sidered: a standard fast-Fourier transform-based Butter worth fi lter (with a 7.5-km half-width) and a space-domain kernel fi lter (with a 9-km half-width), developed by the University of Texas (Holt et al., 2006b). Both fi lters gave similar results along the line. However, there was a signifi cant problem with “ringing” of the Butter worth fi lter close to the end of the lines. The space-domain fi lter was therefore pre-ferred. The fi ltered free-air gravity data were upward continued (Blakely, 1995) to a constant level of 2400 m. The fi nal free-air gravity data were gridded using a tensioned spline technique (Smith and Wessel, 1990) with a tension factor of 0.25 and grid mesh of 5 km (Fig. 3B). The overall quality of the new airborne gravity data is high with a standard deviation for 265 cross-over points of 2.8 mGal. Despite the lack of level ing these crossover errors are comparable to those reported for previous airborne gravity surveys over West and East Antarctica (Bell et al., 1999; Jones et al., 2002; Ferraccioli et al., 2005; Ferraccioli et al., 2006).

We calculated the complete Bouguer correc-tion from the regional topographic grids, which included all available onshore and offshore re-gional data (Holt et al., 2006a; Vaughan et al., 2006). The Bouguer correction was calculated at the 2400 m observation altitude using a 3-D Gauss-Legendre quadrature (GLQ) method (von Frese et al., 1981). The topographic grids included data extending at least 165 km from the gravity fl ights, allowing the computation of a fully terrain-corrected Bouguer gravity anom-aly (i.e., equivalent to Hayford zone O). The gravity effects of ice and water were calculated assuming densities of 915 and 1028 kg m−3, respec tively. The rock (topographic) correction used a standard density of 2670 kg m−3. Bed-rock topography above 0 m (as defi ned by the WGS84 ellipsoid) gave a positive gravity ef-fect and bedrock topography below 0 m gave a negative gravity effect. The summation of the ice, water, and topographic gravity effects gave the total terrain gravity effect. Subtraction of the

Pin

eIslan

dG

lacier

HudsonMountains

EllsworthMountains

Pin

eIslan

dG

lacierHudson

Mountains

H

H

Sinuous

Ridge

Sinuous

Ridge

Bentley Subglacial Trench

Bentley Subglacial Trench

Byrd

Subglacial

Basin

Byrd

Subglacial

Basin

–2250 –900 –500 –250 200 1100 m

Pin

eIslan

dG

lacier

A

HudsonMountains

EllsworthMountains

95°W

105°W

–50 –29 –20 –13 –7 2 17

Pin

eIslan

dG

lacierHudson

Mountains

EllsworthMountains

B

95°W

105°W

76°S 78°S 80°S

mGal

76°S 78°S 80°S

90°W

100°W

90°W

100°W

H

H

Sinuous

Ridge

N

N

Sinuous

Ridge

0 50 10050km

0 50 10050km

Bentley Subglacial Trench

Bentley Subglacial Trench

Byrd

Subglacial

Basin

Byrd

Subglacial

Basin

Figure 3. (A) Subglacial topography beneath the catchment of Pine Island Glacier (brown), modifi ed from Vaughan et al. (2006) and Holt et al. (2006a). Dot-dash line marks major topo graphic discontinuities between subglacial highlands on the southern side of Pine Island Glacier (H) and the Byrd Subglacial Basin, and between the Sinuous Ridge and the Bentley Subglacial Trench. White triangle marks location of passive seismic station (Winberry and Anandakrishnan, 2004). The hashed region at the mouth of Pine Island Glacier, between the ice front and the grounding line, is where bed topography is unconstrained by radar data. (B) Free-air gravity anomaly map with superimposed topographic boundaries. In this and following fi gures a dashed black outline marks the region of fast fl ow associated with Pine Island Glacier. Labels as above.

Aerogravity evidence for major crustal thinning in West Antarctica

Geological Society of America Bulletin, Month/Month 2009 5

26417 4th pages / 5 of 13

terrain gravity effect from the free-air anomaly yielded the complete, terrain-corrected, Bouguer gravity anomaly. The Bouguer anomalies were then gridded using the same technique as the free-air data (Fig. 4A).

AEROGEOPHYSICAL IMAGES

Bedrock Topography and Free-Air Gravity Anomaly Maps

The subice topography (Fig. 3A) and new free-air gravity anomaly (Fig. 3B) show, as expected, a high degree of correlation. The Ellsworth Mountains are over 4000 m high and are associated with positive free-air grav-ity anomalies of up to 60 mGal, in contrast to promi nent free-air gravity lows over the Bentley Subglacial Trench, Byrd Subglacial Basin, and Pine Island Glacier. The Bentley Subglacial Trench is ~2200 m deep and is separated from the broader Byrd Subglacial Basin by a region of elevated, but highly dissected, topography, re-ferred to as Sinuous Ridge (Holt et al., 2006a). Between the Byrd Subglacial Basin and Pine Island Glacier there is also a region of elevated topography, which is less dissected than the Sinuous Ridge (H in Fig. 3A).

Pine Island Glacier fl ows in a confi ned sub-glacial valley up to 1600 m deep. The morphol-ogy of this deep valley suggests that the glacier may fl ow in a graben-like feature. In contrast, the elevated Hudson Mountains lie along the northern fl ank of Pine Island Glacier and feature positive free-air anomalies of up to ~50 mGal. The Hud-son Mountains contain several Cenozoic basaltic volcanoes (LeMasurier and Thomson, 1990).

Bouguer Gravity Anomaly Map

The Bouguer gravity anomaly map (Fig. 4A) reveals a prominent long-wavelength negative anomaly over the Ellsworth Mountains between –20 and –104 mGal, which we interpret as evidence for thick crust. The inferred rift fl ank (Fitzgerald and Stump, 1991) between the Ells-worth Mountains block and the West Antarctic Rift System is marked by a prominent NE-SW oriented Bouguer anomaly gradient (L1 in Fig. 4A). Over the Hudson Mountains the re-gional Bouguer gravity anomaly is between –40 and –10 mGal, which may also refl ect rela-tive crustal thickening. A discrete, ~20 mGal, positive Bouguer anomaly (A) over the Hud-son Mountains is associated with outcrops of Cenozoic basaltic volcanic rocks (LeMasurier and Thomson, 1990) and is likely to arise from dense mafi c intrusions associated with Ceno-zoic to Holocene (Corr and Vaughan, 2008) vol-canism. Regional positive Bouguer anomalies

L1B

A

S S′

G′

G

P P′

AP″

Pin

eIslan

dG

lacier

Hudson

Mountains

Byrd

Subglacial

Basin

Pin

eIslan

dG

lacier

Hudson

Mountains

Byrd

Subglacial

Basin

Sinuous

Ridge

Bentley Subglacial Trench

Sinuous

Ridge

Bentley Subglacial Trench

33188 262–2–7–16–58–90 2214

Pin

eIslan

dG

lacier

Hudson

Mountains

Byrd

Subglacial

Basin

EllsworthMountains

A

90°W

100°W

95°W

105°W

76°S 78°S 80°S

0 50 100km

Pin

eIslan

dG

lacier

Hudson

Mountains

Byrd

Subglacial

Basin

EllsworthMountains

50

BmGal

100°W

76°S 78°S 80°S

–29 –24 –20 –17–15 –13 –11 1–9 –7 –5 –3 4 11 mGal

N

N

21 +/– 1 km

90°W

Sinuous

Ridge

Bentley Subglacial Trench

Sinuous

Ridge

Bentley Subglacial Trench

0 50 10050km

L1B

A

S S′

G′

G

P P′

AP″

95°W

105°W

Figure 4. (A) Bouguer gravity anomaly map. Note data between grounding line and ice front has been excluded because of the lack of bathymetry data. (B) Airy isostatic anomaly map, assuming crustal and mantle densities of 2800 and 3330 kg m−3, respectively. Boxed red number shows Moho depth and associated error estimate recovered at the seismic sta-tion of Winberry and Anandakrishnan (2004), which was used as a reference value to model regional Moho undulations from our new aerogravity data. Letters mark features referred to in the text. Black line marks cross section shown in Figures 5 and 8.

Jordan et al.

6 Geological Society of America Bulletin, Month/Month 2009

26417 4th pages / 6 of 13

overlie the eastern segment of Pine Island Gla-cier, the Byrd Subglacial Basin, and the Bentley Subglacial Trench.

Airy Isostatic Anomaly Map

To provide an initial assessment of the intra-crustal structure across the survey area, we calculated the gravity anomaly due to Airy isostatic compensation of the known surface loads (i.e., subglacial topography and ice thick-ness). To calculate the gravity effect of isostatic compensation a reference Moho depth must be assumed. Using a reference Moho depth of 23 km yields a predicted Airy isostatic Moho in the Bentley Subglacial Trench within 500 m of the independent receiver function estimate (Winberry and Anandakrishnan, 2004).

The Airy isostatic residual anomaly map (Fig. 4B) is obtained by subtracting the Airy model from the Bouguer anomaly. The regional negative Bouguer anomaly over the Ellsworth Mountains is now effectively removed, suggest-ing that this region is close to being in isostatic equilibrium. Negative isostatic residual anoma-lies (S and S′ in Fig. 4B) over the coastal part of Pine Island Glacier and over the Byrd Subglacial Basin may represent the gravity signature of thick sedimentary basins. The interpretation of sedi-mentary infi ll within the Byrd Subglacial Basin is supported by airborne radar data that shows the bedrock in the basin is remarkably smooth. The negative isostatic anomalies within the Ellsworth Mountains (G) and southeast of the Pine Island Glacier (G′) are associated with elevated and rougher topography, and may arise from uplifted low-density sedimentary rocks or granitoids.

Positive Airy isostatic anomalies mark the east-ern end of Pine Island Glacier (P) and the Bentley Subglacial Trench (P′) and could imply thinner crust than predicted by the Airy isostatic model, or dense intrusions associated with subglacial volcanism. The short-wavelength character of the anomalies favors the latter hypothesis. In the Hudson Mountains the positive Bouguer anomaly (A) is enhanced by the Airy isostatic correction to over 28 mGal, supporting our interpretation of a dense mafi c intrusion in this region. Between Pine Island Glacier and the Byrd Subglacial Basin there is a region of small-amplitude positive Airy isostatic residual anomalies (P″), between –2 and 11 mGal, which may also reveal dense rift-related mafi c rocks.

Gravity Signatures along a Profi le across the West Antarctic Rift System

We extracted a cross section across the Pine Island Glacier region to show the relationship between subice topography, free-air gravity, and

Bouguer and Airy isostatic anomalies (Fig. 5). The profi le was chosen so as to cross the main subglacial basins, which we interpret as being part of the West Antarctic Rift System and the fl anking elevated topography of the Hudson and Ellsworth Mountains. In addition, our cross sec-tion intersects the only available seismic station over this part of the West Antarctic Rift System (Winberry and Anandakrishnan, 2004). Positive Bouguer gravity anomalies are detected over Pine Island Glacier, Byrd Subglacial Basin, and the Bentley Subglacial Trench. The Airy anoma-lies show differing signatures over the basins . Both the Pine Island Glacier and the Byrd Sub-glacial Basin exhibit prominent lows, while the Bentley Subglacial Trench features a high. The subice topography (Fig. 5A) suggests that these basins may be graben- or half-graben-like struc-tures and the Airy isostatic signature would be consistent with the presence of sedimentary infi ll in part of Pine Island graben and Byrd Sub glacial Basin. In contrast, the high over the Bentley Sub-glacial Trench would suggest thin sedimentary infi ll, consistent with independent seismic evi-dence for less than 500-m infi ll (Winberry and Anandakrishnan, 2004). To further our under-standing of the source of the positive Bouguer anomalies over the newly identifi ed grabens we estimated the rigidity of the lithosphere and the crustal thickness variations within this part of the West Antarctic Rift System.

ESTIMATING LITHOSPHERE RIGIDITY (Te)

The Airy isostatic model we presented as-sumes that the lithosphere has no lateral strength. However, the response of the lithosphere to loads is better approximated as an elastic plate with an equivalent thickness referred to as Te (Watts, 2001). Because the rigidity of the West Antarctic Rift may have increased with time after rifting, in a similar manner to the oceanic lithosphere (Watts, 2001), estimates of the tim-ing of rifting can also be made by analyzing Te (Karner et al., 2005).

We adopt the method of Stewart and Watts (1997) to assess the elastic thickness for this sector of the West Antarctic Rift System assum-ing that compensation of the ice and topographic loads occurs at the Moho. We determined the best-fi tting regional Te by comparing calculated gravity anomalies from different isostatic mod-els with the observed Bouguer anomaly (Fig. 6). We chose to apply the simplest possible model of isostatic compensation, where load density (ρload) = crustal density (ρcrust) (Stewart and Watts, 1997). We also used parameters com-parable to those of Stewart and Watts (1997), where a Young’s modulus of 1011 Pa and a

Poisson’s ratio of 0.25 defi ne the properties of the elastic beam, and ρload = 2670 kg m−3 and mantle density (ρman) = 3330 kg m−3. The ice thickness and offshore water depth were con-verted into equivalent rock thickness (assuming ρice = 915 kg m−3 and ρwater = 1028 kg m−3) and added to the elevation of the subice topography. Three-dimensional undulations in the isostatic compensating surface were calculated for all grid points simultaneously using a fast Fourier transform routine, grdfft, a component of the

200 400 600 km0

–40

0

40

–40

0

40

–40

40

1.5

–1.5

0

0

HM PIG BSB SR BSTEWMflank

West Antarctic Rift System?AN S

mGal

mGal

mGal

km

B

C

D

SS

Figure 5. Cross-section view across the West Antarctic Rift System underlying the Pine Island Glacier catchment and correspond-ing gravity signatures. EWM—Ellsworth-Whitmore Mountains; SS—seismic station (Winberry and Anandakrishnan, 2004), SR—Sinuous Ridge, other abbreviations as in Figure 1. (A) Subglacial topography of the rift basins and uplifted fl anks. Note the graben-like basin underlying Pine Island Glacier. (B) Free-air gravity anomalies across the rifted region. (C) Bouguer gravity anomalies and (D) Airy isostatic anomalies. See text for detailed description.

Aerogravity evidence for major crustal thinning in West Antarctica

Geological Society of America Bulletin, Month/Month 2009 7

26417 4th pages / 7 of 13

GMT software package (Wessel and Smith, 1991). Isostatic models for constant Te values of between 0 and 40 km were constructed and the gravity fi eld for each model calculated using the GLQ program (von Frese et al., 1981) for a reference Moho depth of 23 km and an observa-tion altitude of 2400 m.

In Figure 6 we analyzed the spatial distribu-tion of residuals calculated as the difference between the observed Bouguer gravity anomaly and the isostatic models for the two end-member models of Te = 0 and Te = 40. The low Te models over this sector of the West Antarctic Rift System yield an improved fi t to the observed data (stan-dard deviation of 10 mGal for Te = 0) compared to the higher Te models (standard deviation of 20 mGal for Te = 40 km). The best-fi t isostatic model was obtained for a Te of (5 ± 5 km), which yielded standard deviations of ~9 mGal.

The Pine Island Glacier graben, the Byrd Subglacial Basin, and the Bentley Subglacial Trench all appear to be underlain by weak rifted lithosphere. The estimated Te value was found to be insensitive to varying the reference Moho depth between 23 and 27 km and crustal load densities between 2670 and 2800 kg m−3. However, since the best reference Moho depth (23 km) occurs at the low end of this range, we performed further tests by decreasing the mini-mum reference Moho depth to 18 km. This was found not to alter the best fi t Te value. Regions of higher rigidity may exist within the West Antarctic Rift System; however, if such higher rigidity blocks are relatively localized and bounded by low rigidity zones, it is likely that the entire region will appear to be in local, rather than regional, isostatic equilibrium.

CRUSTAL THICKNESS ESTIMATES

Crustal thickness variations provide key information about the crustal architecture beneath the Pine Island Glacier catchment,

which may impact overlying ice dynamics. A continental-scale crustal thickness grid for Antarctica (Müller et al., 2007) has been previ-ously derived from the ice thickness and subgla-cial topog raphy compiled as part of BEDMAP (Lythe et al., 2000). However, within the catch-ment of the Pine Island Glacier there is only a single direct estimate of crustal thickness made from receiver function analysis (Winberry and Ananda krishnan, 2004). Our modeling of air-borne gravity data is therefore a new tool to assess crustal thickness variations under the West Antarctic Rift System.

There are two key limitations when recov-ering crustal thickness variations from gravity data. First, intracrustal density contrasts may mask the signal derived from the Moho inter-face. The effect of such intracrustal bodies must therefore be removed before recovery of Moho undulations. Second, modeling of potential fi eld data is non-unique, and assumptions must therefore be made about both the mean depth of the Moho interface and the density contrast across the Moho.

To estimate Moho variations we used three approaches, which are briefl y outlined below; (1) an Airy isostatic model, (2) inversion of the fi ltered Bouguer gravity anomaly, and (3) inversion of the compensating terrain grav-ity effect (Fig. 7).

The Airy isostatic model was based solely on the known surface loads (topography and ice thickness) and was used as a reference for the other models, which include observed grav-ity data. The predicted Moho variations from the Airy isostatic model were low-pass fi ltered (Fig. 7A), as very short wavelength variations in topography are unlikely to be isostatically sup-ported. The fi lter we applied passed wavelengths >75 km, cosine tapered wavelengths between 75 and 60 km, and cut off all wavelengths <60 km.

The Bouguer gravity anomaly at long wave-lengths is dominated by variations in Moho

depth due to isostatic compensation of loads on and within the crust (Watts, 2001). It is there-fore often assumed that inversion of the long-wavelength Bouguer anomaly fi eld provides an estimate of Moho depth (Braitenberg et al., 2000; Studinger and Bell, 2007). We therefore applied a low-pass fi lter (>~75 km for a reference crustal thickness of 35 km) to the Bouguer anomaly to minimize the effect of short-wavelength signals, arising from intracrustal bodies. However, as discussed previously, we assume that a more appropriate reference crustal thickness is 23 km beneath the Pine Island Glacier catchment. This implies that the low-pass fi lter we applied to the Bouguer anomaly data will attenuate part of the signal from the Moho. Therefore we re-gard the amplitude of the recovered Moho undu-la tions as a minimum estimate.

Variations in Moho depth (Fig. 7B) were recovered from the long-wavelength Bouguer gravity anomaly using a 3-D inverse modeling approach based on the GLQ method (von Frese et al., 1981). Initial Moho undulations were in-put and the gravity effect calculated using the GLQ method. The residual between the calcu-lated and observed anomaly was then used to adjust the Moho depth. This process continued iteratively until a good fi t (root-mean-square error <3 mGal) between the calculated and the observed gravity anomaly was achieved.

The third approach we utilized is based on inversion of the compensating terrain gravity effect. The “terrain compensating gravity ef-fect” is the difference between the terrain grav-ity effect and the part of the free-air anomaly that is correlated with the topography. The ter-rain compensating gravity effect is assumed to be due to isostatic compensation of the surface topog raphy at the Moho (Jones et al., 2002; Braun et al., 2007). The decorrelated part of the free-air anomaly is assumed to be due to intra-crustal density variations such as sedimentary basins and igneous intrusions.

To recover the terrain compensating gravity effect we fi rst applied the same low-pass fi l-ter used in the Airy isostatic model to both the terrain gravity effect and the free-air gravity anomaly to minimize short wavelength noise. We then used a spectral correlation fi lter to split the free-air anomaly into two components, one correlated with the topography, and a second decorrelated with the topography. We followed Jones et al. (2002) in assuming that a correlation coeffi cient of 0.64 excludes the majority of the uncorrelated free-air gravity anomaly.

The Moho undulations derived from the compensating terrain gravity effect (Fig. 7C) were recovered using the method stated previ-ously for the long-wavelength Bouguer grav-ity anomaly. Our models assumed a reference

HMBSB SR

BSTHM

BSB SR

BSTHM

BSB

EWM

SR

BSTHM

BSB

EWM

SR

BST

Te = 40Te = 0

150 (km)150 (km)

40A B

–60

–40

–20

0

20

mGal

90°W

76°S

105°W

78°S

90°W

76°S 78°S

105°W

Figure 6. Comparison between end member isostatic models for this part of the West Ant-arctic Rift System. Note the higher amplitude residuals in panel B referring to the higher rigidity model (Te = 40) compared to the Airy model (Te = 0) shown in panel A.

Jordan et al.

8 Geological Society of America Bulletin, Month/Month 2009

26417 4th pages / 8 of 13

Moho depth of 23 km. This yields a crustal thickness of 21 ± 1 km for the Bentley Sub-glacial Trench, which is consistent with the in-dependent receiver function estimate (Winberry and Anandakrishnan, 2004). For the gravity models Moho depths deeper than the reference depth gave a negative anomaly, and shallower depths gave a positive anomaly. A crust/mantle density contrast of 530 kg m−3 and an observa-tion altitude of 2400 m were assumed for all gravity modeling.

Predicted Airy Isostatic Moho Variations

The regional crustal thickness map from the Airy isostatic model suggests crust between 28 and 35 km thick beneath the Ellsworth Mountains and crust over 24 km thick under the Hudson Mountains (Fig. 7A). Narrow linear regions of thinned crust <21 km thick are predicted beneath the Pine Island Glacier graben, Bentley Subglacial Trench, and Byrd Sub glacial Basin.

Moho Variation from Filtered Bouguer Gravity Anomaly

Moho predictions based on the fi ltered Bouguer gravity anomaly exhibit similar pat-terns to the Airy model (compare Figures 7A and 7B), with thicker crust (>30 km) beneath the Ellsworth Mountains and thinned crust be-neath the West Antarctic Rift System. There are, however, regions where the Airy Moho predic-tions and the pattern from the fi ltered Bouguer anomaly differ, for example, beneath the Byrd Subglacial Basin (D in Fig. 7B). Here the Airy isostatic model predicts a more linear region of thinner crust compared to the Bouguer de-rived model. The discrepancy between the Airy isostatic- and Bouguer-derived Moho esti mates may be accounted for if several km of low-density sedimentary infi ll is present in the Byrd Subglacial Basin.

The discrete Moho high (D′) imaged beneath the Hudson Mountains is also likely to be due to contamination of the fi ltered Bouguer anomaly by an intracrustal source, in this case a mafi c intrusion, which is apparent from the Airy iso-static anomaly map (A in Fig. 4B).

Moho Variation from Compensating Terrain Gravity Effect

Depth to Moho estimates recovered from inversion of the compensating terrain gravity effect show an improved match to the predic-tions of the Airy isostatic model, with a more linear region of shallow Moho observed be-neath the Byrd Subglacial Basin (Fig. 7C).

0 50 10050km

PIG

PIG

PIG

PIG

0 50 10050km

0 50 10050km

76°S 80°S

90°W

100°W

76°S 80°S

90°W

100°W

76°S 80°S

90°W

100°W

A

B

C

21 +/– 1 km

20

24

28

22

24

22

20

28

22

22

26

20

28

22

24

24

2226

24

N

N

N

–30

–28

–26

–24

–22

–20

–35km

–18

BST

BST

EllsworthMountains

BST

EllsworthMountains

PIG

EllsworthMountains

PIG

PIG

HM

HM

BSB

BSB

BSB

DD′

D′ H

TR

iftR

iftR

ift D

SinuousRidge

SinuousRidge

SinuousRidge

HM

BST

BST

EllsworthMountains

BST

EllsworthMountains

PIG

EllsworthMountains

PIG

PIG

HM

HM

BSB

BSB

BSB

DD′

D′ H

TR

iftR

iftR

ift D

SinuousRidge

SinuousRidge

SinuousRidge

HM

Figure 7. Maps of crustal thickness variation derived from (A) an Airy isostatic model, (B) in-version of long wavelength Bouguer gravity anomaly fi eld, and (C) inversion of the com-pensating terrain gravity effect. Contour interval = 2 km. Abbreviations: PIG—Pine Island Glacier; BST—Bentley Subglacial Trench; BSB—Byrd Subglacial Basin, HM—Hudson Mountains. Black letters mark features discussed in the text. Other features as in Figure 3A.

Aerogravity evidence for major crustal thinning in West Antarctica

Geological Society of America Bulletin, Month/Month 2009 9

26417 4th pages / 9 of 13

Additionally , the circular shallowing of the Moho in the Hudson Mountains (D′) is attenu-ated, suggesting a lesser degree of intracrustal contamination. The eastern end of the Pine Island Glacier and under the Byrd Subglacial Basin feature the thinnest crust (~19 ± 1 km). The Bentley Subglacial Trench also appears to be associated with a linear region of thinned crust (21 ± 1 km). The Hudson and Ellsworth Mountains crustal thicknesses are >26 and ~35 km, respectively. The crustal thickness is estimated to be between 23 and 26 km beneath the subglacial highlands (H) separating the Pine Island Glacier from the Byrd Subglacial Basin and beneath the Sinuous Ridge.

Comparison of Techniques

Beneath the Byrd Subglacial Basin and the Hudson Mountains, the fi ltered Bouguer anomaly method gave signifi cant variations in the Moho pattern, compared to the Airy iso-static model (compare Figures 7B and 7A). We attribute these discrepancies to incomplete removal of intracrustal sources by the low-pass fi lter. The compensating terrain gravity effect method of Jones et al. (2002) removes intra-crustal contamination more effectively and therefore is our preferred method for estimat-ing Moho depth.

There are, however, limitations to inver-sion of the compensating terrain gravity effect. Intra crustal density contrasts may not always be decorrelated with the surface topography. An example would be sedimentary basins within deep topographic depressions. The cor-relation of low-density sedimentary infi ll with low topog raphy will lead to an underestimate of the compensating terrain gravity effect, and hence an underestimate of the amount of crustal thinning. This may be applicable to the Byrd Subglacial Basin, where we have previously identifi ed the presence of possible sedimentary infi ll. This implies that the crust beneath the Byrd Subglacial Basin may be even thinner than our 19 ± 1 km estimate.

Figure 8 shows a comparison between the dif-ferent model predictions of Moho depth along our sample profi le crossing the West Antarctic Rift System. The most obvious discrepancies between the different Moho depth estimates occur beneath the Pine Island Glacier graben and the Byrd Subglacial Basin. The fi ltered Bouguer gravity method would yield a broad regional Moho upwarp beneath these basins to depths of ~23 km. The inversion of the com-pensating terrain gravity effect (our preferred method) indicates more localized regions of thinner crust (19 ± 1 km). All models converge to a Moho depth of 21 ± 1 km under the Bentley

Subglacial Trench in good agreement with the independent receiver function estimate (Win-berry and Anandakrishnan, 2004).

Sensitivity Tests

The boundary conditions we imposed to ac-count for the non-unique nature of modeling gravity data were a reference Moho depth at 23 km and a density contrast across the Moho of 530 kg m−3. These assumptions will affect the absolute depth and amplitudes of the esti-mated Moho. However, different physical as-

sumptions do not alter the overall pattern of variation. To assess the impact of increasing the reference Moho depth, or varying the den-sity contrast, we ran several sensitivity tests us-ing the Moho derived from the compensating terrain gravity effect as a reference (Table 1). Increasing the reference Moho depth from 23 to 27 km increases the magnitude of the esti-mated Moho undulations from ~16 to ~17 km across the entire region. Decreasing the den-sity contrast from 530 to 430 kg m−3 had a more signifi cant impact, yielding maximum Moho depths 3 km deeper and minimum Moho

2

–2

0

–22

–20

–26

–28

–24

100 200 300 400 500 600 700

SeismicStation

km

km

0

20

–20

–40

–60

mGal

HM PIG BSB SR BSTEWMflank

West Antarctic Rift System?

A

Airy MohoTerrain compensating MohoBouguer Moho

Terrain compensating gravity effectFiltered Bouguer anomaly

Ice

Bedrock

B

DD′

N S

–18

Figure 8. Comparison between crustal thickness estimates derived from the Airy, Bouguer, and terrain compensating methods along the sample profi le crossing the West Antarctic Rift System. Panel A shows the input low-pass fi ltered Bouguer gravity anomaly along the profi le and the terrain compensating gravity effect. Panel B shows crustal thickness esti-mates under the different rift basins. The gray envelope denotes the range of crustal thick-ness estimates obtained by using the terrain compensating method with different crust/mantle density contrasts (see Table 1). Also note the error bar associated with the receiver function estimate for crustal thickness under the Byrd Subglacial Trench (Winberry and Anandakrishnan, 2004).

Jordan et al.

10 Geological Society of America Bulletin, Month/Month 2009

26417 4th pages / 10 of 13

depths 1 km shallower. In all these tests, the pattern of Moho variation about the reference level remained the same.

DISCUSSION

Crustal Boundaries

The new airborne gravity data collected across the catchment of Pine Island Glacier provides new insights into the crustal structure of the West Antarctic Rift System. The tectonic boundary between the West Antarctic Rift Sys-tem and the Ellsworth Mountains block is now clearly imaged from the new Bouguer anomaly map (L1 in Fig. 4A). Our interpretation for the Ellsworth Mountains as the uplifted fl ank of the West Antarctic Rift System would be com-patible with apatite-fi ssion track data, which show ~4 km of uplift since the Early Cretaceous (Fitzgerald and Stump, 1991). Crustal thickness beneath the Ellsworth Mountains, as recovered by inversion of the compensating terrain grav-ity effect, is ~35 km (Fig. 7C). Therefore the crust beneath the Ellsworth Mountains rift fl ank exhibits comparable thickness to the Whitmore Mountains further to the west (Bell et al., 1999; Studinger et al., 2002) and to the Transantarctic Mountains adjacent to the Ross Sea Rift (Della Vedova et al., 1997; Ferraccioli et al., 2001).

The margin between the West Antarctic Rift System and the Hudson Mountains is less dis-tinct in the new Bouguer gravity anomaly map. The crustal thickness variations recovered from the inversion of the compensating terrain gravity effect (Fig. 7C) reveal a narrow rift zone under-lying graben-like topography of the Pine Island Glacier. We name this feature the Pine Island Rift. It is possible that the Pine Island Rift exploited an older inherited tectonic structure, perhaps the southern boundary of the Thurston Island block (Gohl et al., 2007). However, aerogravity data coverage over the Thurston Island block would be necessary to confi rm or refute this hypothesis.

Crustal Thinning

Modeling of the new airborne gravity data shows that the crust between the Ellsworth and Hudson Mountains is signifi cantly thinned,

indicating that the West Antarctic Rift System extends beneath the catchment of Pine Island Glacier. Predicted Moho depths are between ~26 and ~19 km. The subglacial rifts we im-aged follow two distinct trends. The Bentley Subglacial Trench is NE-SW oriented and lies parallel to the Ellsworth Mountains rift fl ank. In contrast, both the Pine Island Rift and the Byrd Subglacial Basin are approximately E-W ori-ented and hence lie at a high angle to the West Antarctic Rift fl ank.

Crustal thickness beneath the catchment of Pine Island Glacier is only slightly higher than the Ross Sea Rift region, where values between 17 and 26 km have been reported (Trey et al., 1999; Bannister et al., 2003). However, sedi-ment thickness in the Ross Sea Rift can reach up to 14 km in the Victoria Land Basin (Branco-lini et al., 1995). A comparable thickness of sediments is not present beneath the Pine Island Rift, Bentley Subglacial Basin, or Bentley Sub-glacial Trench based upon initial depth to mag-netic basement estimates (Ferraccioli, 2008, personal commun.), or the single sediment thickness estimate of 550 ± 250 m, derived from receiver function analysis (Anandakrishnan and Winberry, 2004). Between the regions of highly thinned crust we model Moho depths of ~26 km (e.g., beneath the northern part of the Sinuous Ridge), which is similar to the estimated aver-age crustal thickness derived from analysis of airborne gravity data over the West Antarctic Rift System in the Siple Coast region (Studinger et al., 2002). In the northeastern corner of the survey area, a region of thinned crust (T in Fig. 7C) suggests that the West Antarctic Rift System may continue from the Amundsen Sea Embayment toward the Bellingshausen Sea.

Crustal thinning beneath the catchment of the Pine Island Glacier is highly signifi cant since it suggests that elevated geothermal heat fl ux may occur. If a rifting event occurred relatively re-cently (<ca. 30 Ma), elevated heat fl ux would be caused by thinning of the nonconvecting thermal lithosphere (McKenzie, 1967; Nissen et al., 1995) and advection of hot mantle-derived magma into the crust. Mantle-derived magma is likely to give rise to mafi c intracrustal intru-sions, which would create positive Airy isostatic gravity anomalies such as those we observe

(P and P″ in Fig. 4B). Enhanced heat fl ux asso-ciated with crustal thinning processes would be in addition to the regional increase associated with a possible mantle plume beneath the West Antarctic Rift System (Behrendt et al., 1996). Elevated heat fl ux could provide a source of enhanced basal melting beneath the Amundsen Sea Embayment sector of the West Antarctic Ice Sheet, which would impact on ice dynamics and hence long-term ice-sheet stability (Fahnestock et al., 2001; Parizek et al., 2002).

Crustal Thickness Variations and Inferred Rifting Stages

The range in crustal thicknesses (~19 to ~26 km) we observed under the West Antarctic Rift System between the Ellsworth Mountains and the Hudson Mountains may be explained by inferring two different rifting phases. In this scenario, initial distributed rifting, in the Creta-ceous, thinned the crust from ~35 to ~26 km. A later narrower phase of rifting, which we specu-late to be Cenozoic in age, caused more local-ized crustal extension beneath the Pine Island Rift and Byrd Subglacial Basin, where the crust is <20 km thick.

Our gravity-derived interpretation for a two-stage rifting process under the catchment of Pine Island Glacier would be compatible with previ-ous geophysical interpretations for multistage rifting within the Ross Sea segment of the West Antarctic Rift System (Cooper and Davey, 1985; Davey and Brancolini, 1995). Notably, several stages of rifting are observed in many other major continental rift systems worldwide (Olsen, 1995), and different modes of rifting and strain localization (i.e., transitions from wide mode to narrow mode) have been modeled (Benes and Davey, 1996). A two-stage rifting scenario is also compatible with a recent numeri cal model of rift localization put forward for the Ross Sea Rift region, where a change from distributed to localized rifting within the Victoria Land Basin occurs (Huerta and Harry, 2007).

If a Cenozoic phase of rifting affected the Amundsen Sea Embayment, it may specu-latively be linked with spreading within the Adare Trough, which was active between 43 and 26 Ma (Cande et al., 2000). Opening of the Adare Trough may have induced distributed ex-tension within the Ross Sea segment of the West Antarctic Rift System (Decesari et al., 2007). Rotation about the pole suggested by Davey et al. (2006) would induce orthogonal rifting in the Victoria Land Basin, i.e., subparallel to the Transantarctic Mountains, but rifting oblique to the Ellsworth Mountains further to the south. An oblique Cenozoic rifting process may ex-plain the E-W trend of the Pine Island and Byrd

TABLE 1. MINIMUM AND MAXIMUM ESTIMATED MOHO DEPTHS, BASED ON INVERSION OF THECOMPENSATING TERRAIN GRAVITY EFFECT, FOR VARIOUS REFERENCE CRUSTAL THICKNESSESAND DENSITY CONTRASTS ACROSS THE MOHO. BOLD NUMBERS ARE OUR PREFERRED VALUES

Reference crustal thickness (km)

Density contrast (kg m−3)

Maximum depth (km)

Minimum depth (km)

Range (km)

03432 –37.7 –17.7 20.0 03532 –34.7 –18.7 16.1 06632 –32.3 –19.5 12.8 03572 –40.8 –23.4 17.3

Aerogravity evidence for major crustal thinning in West Antarctica

Geological Society of America Bulletin, Month/Month 2009 11

26417 4th pages / 11 of 13

Subglacial Basin rifts, which both run at ~45° to the Ellsworth Mountains rift fl ank. Alterna-tively, or additionally, Cenozoic dextral strike-slip motion could have induced oblique rifting beneath the Amundsen Sea segment of the West Antarctic Rift System (Müller et al., 2007).

Te and Age of Rifting

The observed positive Bouguer gravity anom-alies, associated with deep basins within the Pine Island Glacier catchment, are similar in pattern to those detected over the Central Basin of the Ross Sea Rift (Karner et al., 2005). The positive Bouguer anomalies over the Central Basin of the Ross Sea Rift have been explained by postulat-ing that there is a time lag between initial Cre-taceous rifting, when the lithosphere was weak (Te ~0 km), and later Cenozoic sedimentation when the lithosphere had regained its rigidity (Te ~30 km) (Karner et al., 2005). The increase in rigidity reduces the amount of subsidence at the Moho associated with isostatic compensation of the sedimentary load. The same model has been applied to the Siple Coast region, where positive Bouguer gravity anomaly signatures over the West Antarctic Rift System are associated with major sedimentary basins (Bell et al., 2006). However, Decesari et al. (2007) used subsidence modeling over the Central Basin of the Ross Sea Rift to infer ~40 km of Tertiary extension, thereby questioning the lack of post-Cretaceous extension inferred by Karner et al. (2005).

Under the Pine Island Glacier catchment we have recovered a low rigidity (Te of 5 ± 5 km) (Fig. 5). Therefore in contrast to previous in-terpretations over both the Ross Sea and Siple Coast regions, we interpret the positive Bouguer anomalies as arising from isostatic compensa-tion of bedrock topography and ice load. The low Te over the Amundsen Sea Embayment re-gion could refl ect a Cenozoic rifting stage. Al-ternatively the present-day topography and low Te could refl ect a “fossil” isostatic response to initial Cretaceous rifting. However, the devel-opment of a low-relief erosion surface, which has been subsequently uplifted and faulted (LeMasurier and Landis, 1996) argues against the alternative fossil Cretaceous interpretation. The Byrd Subglacial Basin and Bentley Sub-glacial Trench have recently been interpreted to be young graben-like structures of inferred Neo-gene age (LeMasurier, 2008). The low Te values we obtain under the catchment of Pine Island Glacier are similar to other estimates of Te in ac-tive continental rift zones such as ~5 km in the Basin and Range Province (Lowry and Smith, 1994) and are slightly lower than the minimum estimates of ~14 km from the East African Rift system (Tessema and Antoine, 2003).

CONCLUSIONS

We have presented the fi rst airborne gravity images over the catchment of Pine Island Gla-cier coupled with gravity modeling results. Our analysis provides several new insights into the crustal structure beneath this dynamic part of the West Antarctic Ice Sheet.

(1) The crust under the catchment of Pine Is-land Glacier has been signifi cantly thinned as a result of continental rifting associated with the West Antarctic Rift System. Crustal thickness is ~35 km under the Ellsworth Mountains rift fl ank. In contrast, the crust within the rift is be-tween ~26 and 19 km thick. The Byrd Sub glacial Basin and the newly identifi ed Pine Island Rift feature the thinnest crust (~19 ± 1 km) observed beneath the West Antarctic Ice Sheet.

(2) Lithospheric rigidity is low (Te of 5 ± 5 km) under this segment of the West Ant-arctic Rift System. The low Te supports the inter-pretation for an inferred Cenozoic rifting stage that reactivated this segment of the West Antarc-tic Rift System.

(3) We interpret multiple stages of rifting be-neath the Amundsen Sea Embayment. Our inter-pretation predicts Cretaceous wide mode rifting followed by narrow mode rifting in the Ceno-zoic. We infer that Cenozoic rifting may provide a source of enhanced heat fl ow, potentially af-fecting the dynamics of Pine Island Glacier.

(4) The Byrd Subglacial Basin and the west-ern part of the Pine Island Rift are interpreted to be sedimentary basins, which may enhance glacial fl ow in the catchment of Pine Island Glacier. In addition we have inferred Cenozoic mafi c intrusions, which could represent a local source for elevated heat fl ow in the Pine Island Glacier region.

ACKNOWLEDGMENTS

We acknowledge support and funding from the Long-Term Survey and Monitoring Programme of the Geological Sciences Division (BAS). We grate-fully acknowledge reviewers Wesley LeMasurier and an anonymous reviewer for their constructive advice and their corrections that helped improve the paper signifi cantly. We also thank the associate editor for further suggestions. This paper is a contribution to IPY project (ID No: 107) Geodynamics of the West Antarctic Rift System (WARS) in Remote Ellsworth Land and its implications for the stability of the West Antarctic Ice Sheet. We also wish to acknowledge National Science Foundation grant OPP-0230197 for supporting the fi eldwork. We also wish to thank P.T. Leat (BAS) for his advice and comments, which improved upon a draft manuscript.

REFERENCES CITED

Anandakrishnan, S., and Winberry, J.P., 2004, Antarctic subglacial sedimentary layer thickness from receiver function analysis: Global and Planetary Change, v. 42, p. 167–176, doi: 10.1016/j.gloplacha.2003.10.005.

Bannister, S., Yu, J., Leitner, B., and Kennett, B.L.N., 2003, Variations in crustal structure across the transition from West to East Antarctica, Southern Victoria Land: Geo-physical Journal International, v. 155, p. 870–884, doi: 10.1111/j.1365-246X.2003.02094.x.

Behrendt, J.C., 1999, Crustal and lithospheric structure of the West Antarctic Rift System from geophysical in-vestigations - a review: Global and Planetary Change, v. 23, p. 25–44, doi: 10.1016/S0921-8181(99)00049-1.

Behrendt, J.C., Saltus, R., Damaske, D., McCaffrey, A., Finn, C., Blankenship, D.D., and Bell, R., 1996, Pat-terns of Late Cenozoic volcanic and tectonic activity in the West Antarctic Rift System revealed by aero-magnetic surveys: Tectonics, v. 15, p. 660–676, doi: 10.1029/95TC03500.

Bell, R., Blankenship, D.D., Finn, C.A., Morse, D.L., Scambos , T.A., Brozena, J.M., and Hodge, S.M., 1998, Infl uence of subglacial geology on the onset of a West Antarctic ice stream from aerogeophysical observa-tions: Nature, v. 394, p. 58–62, doi: 10.1038/27883.

Bell, R.E., Childers, V.A., Arko, R.A., Blankenship, D.D., and Brozena, J.M., 1999, Airborne gravity and precise positioning for geologic applications: Journal of Geo-physical Research, v. 104, no. B7, p. 15,281–17,292, doi: 10.1029/1999JB900122.

Bell, R., Studinger, M., Karner, G.D., Finn, C., and Blanken-ship, D., 2006, Identifying major sedimentary basins beneath the West Antarctic Ice Sheet from aeromag-netic data analysis, in Fütterer, D.K., et al., eds., Ant-arctica: Contributions to Global Earth Sciences: Berlin, Springer, p. 117–121.

Benes, V., and Davey, P., 1996, Modes of continental litho-spheric extension: Experimental verifi cation of strain localization processes: Earth and Planetary Science Letters, v. 254, p. 69–87.

Blakely, R.J., 1995, Potential Theory in Gravity and Mag-netic Applications: Cambridge, Cambridge University Press.

Blankenship, D.D., Bell, R.E., Hodge, S.M., Brozena, J.M., Behrendt, J.C., and Finn, C.A., 1993, Active volcanism beneath the West Antarctic ice sheet and implications for ice-sheet stability: Nature, v. 361, p. 526–529, doi: 10.1038/361526a0.

Braitenberg, C., Zadro, M., Fang, J., Wang, Y., and Hsu, H.T., 2000, The gravity and isostatic Moho undulations in Qinghai-Tibet plateau: Journal of Geodynamics, v. 30, p. 489–505, doi: 10.1016/S0264-3707(00)00004-1.

Brancolini, G., Busetti, M., Marchetti, A., De Santis, L., Zanolla, C., Cooper, A.K., Cochrane, G.R., Zayatz, I., Belyaev, V., Knyazev, M., Vinnikovskaya, O., Davey, F., and Hinz, K., 1995, Descriptive text for the seis-mic stratigraphic atlas of the Ross Sea, Antarctica, in Cooper, A.K., et al., eds., Geology and Seismic Stra-tigraphy of the Antarctic Margin: Washington, D.C., American Geophysical Union.

Braun, A., Rae Kim, H., Csatho, B., and Von Frese, R.R.B., 2007, Gravity-inferred crustal thickness of Greenland: Earth and Planetary Science Letters, v. 262, p. 138–158, doi: 10.1016/j.epsl.2007.07.050.

Cande, S.C., Stock, J.M., Müller, R.M., and Ishihara, T., 2000, Cenozoic motion between East and West Ant-arctica: Nature, v. 404, p. 145–150, doi: 10.1038/35004501.

Cape-Roberts-Science-Team, 2000, Studies from the Cape Roberts Project, Ross Sea, Antarctica, Terra Antartica, p. 1–654.

Cooper, A.K., and Davey, F.J., 1985, Marine geological and geophysical investigations in the Ross Sea, Antarctica: Antarctic Journal of the United States, v. 19, p. 80–82.

Corr, H., and Vaughan, D.G., 2008, A recent volcanic erup-tion beneath the West Antarctic ice sheet: Nature Geo-science, v. 1, p. 122–125, doi:10.1038/ngeo106.

Dalziel, I.W.D., 2006, On the extent of the active West Ant-arctic Rift System: Terra Antarctica, v. 12, p. 193–202.

Dalziel, I.W.D., and Elliot, D.H., 1982, West Antarctica: Problem child of Gondwanaland: Tectonics, v. 1, p. 3–19.

Dalziel, I.W.D., and Lawver, L., 2001, The lithospheric set-ting of the West Antarctic Ice Sheet, in Alley, R.B., et al., eds., The West Antarctic Ice Sheet: Behavior and Environment, Washington, D.C., American Geophysi-cal Union, p. 29–44.

Jordan et al.

12 Geological Society of America Bulletin, Month/Month 2009

26417 4th pages / 12 of 13

Damaske, D., Behrendt, J.C., McCaffrey, A., Saltus, R., and Meyer, U., 1994, Transfer faults in the west-ern Ross Sea: New evidence from McMurdo Sound/Ross ice shelf aeromagnetic survey (GANOVEX VI): Antarctic Science, v. 6, p. 359–364, doi: 10.1017/S0954102094000556.

Davey, F.J., and Brancolini, G., 1995, The late Mesozoic and Cenozoic Structural setting of the Ross Sea Region, in Cooper, A.K., et al., eds., Geology and Seismic Stra-tigraphy of the Antarctic Margin: Antarctic Research Series : Washington, D.C., American Geophysical Union, p. 167–182.

Davey, F.J., Cande, S.C., and Stock, J.M., 2006, Extension in the western Ross Sea region-links between Adare Basin and Victoria Land Basin: Geophysical Research Letters, v. 33, 5 p., doi: 10.1029/2006GL027383.

Davis, C.H., Li, Y., McConnell, J.R., Frey, M.M., and Hanna, E., 2005, Snowfall-driven growth in East Antarctic Ice Sheet mitigates recent sea-level rise: Science, v. 308, p. 1898–1901, doi: 10.1126/science.1110662.

Decesari, R.C., Wilson, D.S., Luyendyk, B.P., and Faulkner, M., 2007, Cretaceous and Tertiary extension through-out the Ross Sea, Antarctica, in Cooper, A.K., et al., eds., A Keystone in a Changing World – Online Pro-ceedings of the 10th ISAES, Short Research Paper 098, 6 p., doi: 0.3 33/of2007–1047.srp098.

Della Vedova, B., Pellis, G., Trey, H., Zhang, J., Cooper, A.K., Makris, J., and the-ACRUP-working-group, 1997, Crustal structure of the Transantarctic Moun-tains, Western Ross Sea, in Ricci, C.A., ed., The Ant-arctic Region: Geological evolution and processes: Siena, Terra Antartica Publication, p. 609–618.

Diehl, T.M., Holt, J.W., Blankenship, D.D., Young, D.A., Jordan, T.A., and Ferraccioli, F., 2008, First airborne gravity results over the Thwaites Glacier Catchment, West Antarctica: Geochemistry, Geophysics, Geosys-tems, 4 p., doi: 10.1029/2007GC001878 (in press).

Fahnestock, M., Abdalati, W., Joughin, I., Brozena, J., and Gogineni, P., 2001, High geothermal heat fl ow, basal melt, and the origin of rapid ice fl ow in Central Green-land: Science, v. 294, p. 2338–2342, doi: 10.1126/science.1065370.

Ferraccioli, F., Armadillo, A., Bozzo, E., and Privitera, E., 2000, Magnetics and gravity image tectonic frame-work of the Mount Melbourne Volcano area (Antarc-tica): Physics and Chemistry of the Earth, v. 25, no. 4, p. 387–393, doi: 10.1016/S1464-1895(00)00061-2.

Ferraccioli, F., Coren, F., Bozzo, E., Zanolla, C., Gandolfi , S., Tabacco, I.E., and Frezzotti, M., 2001, Rifted crust at the East Antarctic Craton margin; gravity and magnetic interpretation along a traverse across the Wilkes subglacial basin region: Earth and Planetary Science Letters, v. 192, p. 407–421, doi: 10.1016/S0012-821X(01)00459-9.

Ferraccioli, F., Bozzo, E., and Damaske, D., 2002, Aeromag-netic signatures over western Marie Byrd Land provide insight into magmatic arc basement, mafi c magma-tism and structure of the Eastern Ross Sea Rift fl ank: Tectonophysics, v. 347, p. 139–165, doi: 10.1016/S0040-1951(01)00242-6.

Ferraccioli, F., Jones, P.C., Curtis, M.L., and Leat, P.T., 2005, Subglacial imprints of early Gondwana break-up as identifi ed from high resolution aerogeophysical data over western Dronning Maud Land, East Ant-arctica: Terra Nova, v. 17, p. 573–579, doi: 10.1111/j.1365-3121.2005.00651.x.

Ferraccioli, F., Jones, P.C., Vaughan, A.P.M., and Leat, P.T., 2006, New aerogeophysical view of the Ant-arctic Penin sula: More pieces, less puzzle: Geo-physical Research Letters, v. 33, 4 p., doi: 10.1029/2005GL024636.

Ferraccioli, F., Jordan, T.A., Vaughan, D.G., Holt, J.W., James, P.R., Corr, H., Blankenship, D.D., Fairhead, J.D., and Diehl, T.M., 2007, New aerogeophysical sur-vey targets the extent of the West Antarctic Rift Sys-tem over Ellsworth Land, in Cooper, A.K., et al., eds., Antarctica: A Keystone in a Changing World –Online Proceedings of the 10th ISAES X, USGS, p. 4.

Fielding, C., Henrys, S.A., and Wilson, T.J., 2006, Rift his-tory of the western Victoria Land Basin: A new per-spective based on integration of cores with seismic refl ection data, in Futterer, D.K., et al., eds., Antarc-

tica: Contributions to Global Earth Sciences: Berlin, Springer-Verlag, p. 307–316.

Finn, C., Dietmar-Müller, R., and Panter, K.S., 2005, A Ceno zoic diffuse alkaline magmatic province (DAMP) in the southwest Pacifi c without rift or plume origin: Geochemistry, Geophysics, Geosystems, v. 6, 26 p., doi: 10.1029/2004GC000723.

Fitzgerald, P.G., and Stump, E., 1991, Early Cretaceous uplift in the Ellsworth Mountains of West Ant-arctica: Science, v. 254, p. 92–94, doi: 10.1126/science.254.5028.92.

Gohl, K., Tererin, D., Eagles, G., Netzeband, G., Grobys, J.W.G., Parsiegla, N., Schlüter, P., Leinweber, V., Larter , R.D., Uenzelmann-Neben, G., and Udintsev , G.B., 2007, Geophysical survey reveals tectonic structures in the Amundsen Sea embayment, West Antarctica, in Cooper, A.K., et al., eds., Antarctica: A Keystone in a Changing World – Online Proceed-ings of the 10th ISAES, USGS Open-File Report, doi:10.3133/of2007-1047.srp047.

Harlan, R.B., 1968, Eotvos corrections for airborne gravity: Journal of Geophysical Research, v. 73, p. 4675–4679, doi: 10.1029/JB073i014p04675.

Holt, J.W., Blankenship, D.D., Morse, D.L., Young, D.A., Peters, M.E., Kempf, S.D., Richter, T.G., Vaughan, A.P.M., and Corr, H., 2006a, New boundary condi-tions for the West Antarctic Ice Sheet: Subglacial topography of the Thwaites and Smith Glacier Catch-ments: Geophysical Research Letters, v. 33, 4 p., doi: 10.1029/2005GL025561.

Holt, J.W., Richter, T.G., Kempf, S.D., and Morse, D.L., 2006b, Airborne gravity over Lake Vostok and ad-jacent highlands of East Antarctica: Geochemistry, Geophysics, Geosystems, v. 7, no. 11, 15 p., doi: 10.1029/2005GC001177.

Huerta, A.D., and Harry, D.L., 2007, The transition from diffuse to focused extension: Modeled evolution of the West Antarctic Rift system: Earth and Planetary Science Letters, v. 255, p. 133–147, doi: 10.1016/j.epsl.2006.12.011.

Hulbe, C.L., and MacAyeal, D.R., 1999, A new numerical model of coupled inland ice sheet, ice stream, and ice shelf fl ow and its application to the West Antarctic Ice Sheet: Journal of Geophysical Research, v. 104, p. 349–366, doi: 10.1029/1999JB900264.

Jones, P.C., and Ferris, J.K., 1999, Short note: A new esti-mate of the adopted gravity value at Rothera Station, Antarctic Peninsula: Antarctic Science, v. 11, p. 461–462, doi: 10.1017/S0954102099000590.

Jones, P.C., and Johnson, A.C., 1995, Airborne gravity survey in southern Palmer Land, Antarctica, in Pro-ceedings of IAG Symposium on Airborne Field De-termination, IUGG XXI General Assembly, Boulder, Colorado, p. 117–123.

Jones, P.C., Johnson, A.C., von Frese, R.R.B., and Corr, H., 2002, Detecting rift basins in the Evans Ice Stream region of West Antarctica using airborne gravity data: Tectonophysics, v. 347, p. 25–41, doi: 10.1016/S0040-1951(01)00236-0.

Jordan, T.A., Ferraccioli, F., Corr, H., Robinson, C., Caneva, G., Armadillo, A., Bozzo, E., and Frearson, N., 2007, Linking the Wilkes Subglacial Basin, the Transantarc-tic Mountains, and the Ross Sea with a new airborne gravity survey: Terra Antartica Reports, v. 13, p. 18.

Karner, G.D., Studinger, M., and Bell, R., 2005, Gravity anomalies of sedimentary basins and their mechanical implications: Application to the Ross Sea basins, West Antarctica: Earth and Planetary Science Letters, v. 235, p. 577–596, doi: 10.1016/j.epsl.2005.04.016.

LaCoste, L.J.B., 1967, Measurement of gravity at sea and in the air: Reviews of Geophysics, v. 5, no. 4, p. 477–526, doi: 10.1029/RG005i004p00477.

Larter, R.D., Cunningham, A.P., and Barker, P.F., 2002, Tectonic evolution of the Pacifi c margin of Antarctica 1. Late Cretaceous tectonic reconstructions: Journal of Geophysical Research, v. 107, no. B12, 19 p., doi: 10.1029/2000JB000052.

Lawver, L., and Gahagan, L.M., 1994, Constraints on tim-ing of extension in the Ross Sea region: Terra Antartica Reports, v. 1, p. 545–552.

LeMasurier, W.E., 2008, Neogene extension and basin deepening in the West Antarctic rift inferred from com-

parisons with the East African rift and other analogs: Geology, v. 36, p. 247–250, doi: 10.1130/G24363A.1.

LeMasurier, W.E., and Landis, C.A., 1996, Mantle-plume activity recorded by low-relief erosion surfaces in West Antarctica and New Zealand: Geological Society of America Bulletin, v. 108, p. 1450–1466, doi: 10.1130/0016-7606(1996)108<1450:MPARBL>2.3.CO;2.

LeMasurier, W.E., and Thomson, J.W., 1990, Volcanoes of the Antarctic Plate and Southern Oceans, Antarctic Re-search Series: Washington, D.C., American Geophysi-cal Union, p. 512.

Liu, H., Jezek, K., Li, B., and Zhao, Z., 2001, Radarsat Antarctic Mapping Project digital elevation model ver-sion 2: Boulder, Colorado, National Snow and Ice Data Center, digital media.

Lowry, A.R., and Smith, R.B., 1994, Flexural rigidity of the Basin and Range-Colorado Plateau-Rocky Moun-tains transition from coherence analysis of gravity and topog raphy: Journal of Geophysical Research, v. 99, p. 20,123–20,140, doi: 10.1029/94JB00960.

Luyendyk, B.P., 1995, Hypothesis for Cretaceous rifting of east Gondwana caused by subducted slab capture: Geology, v. 23, p. 373–376, doi: 10.1130/0091-7613(1995)023<0373:HFCROE>2.3.CO;2.

Lythe, M.B., Vaughan, D.G., and the-BEDMAP-Consor-tium, 2000, BEDMAP – bed topography of the Antarc-tic: British Antarctic Survey.

Mader, G.L., 1992, Rapid static and kinematic global posi-tioning system solutions using the ambiguity function technique: Journal of Geophysical Research, v. 97, p. 3271–3283, doi: 10.1029/91JB02845.

McKenzie, D.P., 1967, Some remarks on heat fl ow and grav-ity anomalies: Journal of Geophysical Research, v. 72, p. 6261–6271, doi: 10.1029/JZ072i024p06261.

Mukasa, S.B., and Dalziel, I.W.D., 2000, Marie Byrd Land, West Antarctica: Evolution of Gondwana’s Pacifi c margin constrained by zircon U-Pb geochronology and feldspar common-Pb isotopic compositions: Geologi-cal Society of America Bulletin, v. 112, p. 611–627, doi: 10.1130/0016-7606(2000)112<0611:MBLWAE>2.3.CO;2.

Müller, R.M., Cande, S.C., Stock, J.M., and Keller, W.R., 2005, Crustal structure and rift fl ank uplift of the Adare Trough, Antarctica: Geochemistry, Geophysics, Geo-systems, v. 6, 16 p., doi: 10.1029/2005GC001027.

Müller, R.M., Gohl, K., Cande, S.C., Goncharov, A., and Golynsky, A.V., 2007, Eocene to Miocene geom-etry of the West Antarctic Rift System: Australian Journal of Earth Sciences, v. 54, p. 1033–1045, doi: 10.1080/08120090701615691.

Nissen, S.S., Hayer, D.E., Bochu, Y., Weijun, Z., Yongqin, C., and Xiaupin, X., 1995, Gravity, heat fl ow, and seis-mic constraints on the processes of crustal extension: Northern margin of the South China Sea: Journal of Geophysical Research, v. 100, p. 22,447–22,483, doi: 10.1029/95JB01868.

Olsen, K.H., 1995, Continental rifts: Evolution, structure, tectonics (Developments in geotectonics), New York, Elsevier Science, 490 p.

Parizek, B.R., Alley, R.B., Anandakrishnan, S., and Conway, H., 2002, Subcatchment melt and long term stability of ice stream D, West Antarctica: Geophysical Research Letters, v. 29, 4 p., doi: 10.1029/2001GL014326.

Rignot, E.J., 1998, Fast recession of a West Antarctic Glacier: Science, v. 281, p. 549–551, doi: 10.1126/science.281.5376.549.

Rignot, E.J., Thomas, R., Kanagaratnam, P., Casassa, G., Frederick, E., Gogineni, S., Krabill, W., Rivera, A., Russell, R., Sonntag, J., Swift, R., and Yungel, J., 2004, Improved estimation of the mass balance of glaciers draining into the Amundsen Sea sec-tor of West Antarctica from the CECS/NASA 2002 campaign: Annals of Glaciology, v. 39, p. 231–237, doi: 10.3189/172756404781813916.

Rignot, E.J., Bamber, J.L., Van Den Broeke, M.R., Davis, C., Li, Y., Van De Berg, W.J., and Van Meijgaard, E., 2008, Recent Antarctic ice mass loss from radar inter-ferometry and regional climate modelling: Nature Geo-science, v. 1, p. 106–110, doi: 10.1038/ngeo102.

Rocchi, S., Armienti, P., D’Orazio, M., Tonarini, S., Wijbrans , J.R., and Di Vincenzo, G., 2002, Cenozoic magmatism in the western Ross Embayment: Role of mantle plume

Aerogravity evidence for major crustal thinning in West Antarctica

Geological Society of America Bulletin, Month/Month 2009 13

26417 4th pages / 13 of 13

versus plate dynamics in the development of the West Antarctic Rift System: Journal of Geophysical Re-search, v. 107, 22 p., doi:10.1029/2001JB000515.

Rocchi, S., LeMasurier, W.E., and Di Vincenzo, G., 2006, Oligocene to Holocene erosion and glacial history in Marie Byrd Land, West Antarctica, inferred from exhu-mation of the Dorrel Rock intrusive complex and from volcano morphologies: GSA Bulletin, v. 118, doi: 10.1130/B25675.1.

Sasagawa, G., Meunier, T.K., Mullins, J.L., McAdoo, D., and Klopping, F., 2004, Absolute Gravimetry in Antarctica: 1995 observations at McMurdo Station and Terra Nova Bay Station: Open-File report 2004–1190, 49 p.

Shepherd, A., Wingham, D., and Rignot, E.J., 2004, Warm ocean is eroding West Antarctic Ice Sheet: Geo-physical Research Letters, v. 31, 4 p., doi: 10.1029/2004GL021106.

Siddoway, C.S., Richard, S., Fanning, C.M., and Luyendyk, B.P., 2004, Origin and emplacement mechanisms for a middle Cretaceous gneiss dome, Fosdick Mountains, West Antarctica, in Whitney, D.L., et al., eds., Gneiss domes in orogeny, Geological Society of America Spe-cial Paper 380, p. 267–294.

Siddoway, C.S., Sass, L.C.I., and Esser, R., 2005, Kinematic history of the Marie Byrd Land terrane, West Antarc-tica: Direct evidence from Cretaceous mafi c dykes, in Vaughan, A.P.M., et al., eds., Terrane processes at the margin of Gondwana: Special Publication Geological Society of London 246, p. 417–438.

Smith, W.H.F., and Wessel, P., 1990, Gridding with con-tinuous curvature splines in tension: Geophysics, v. 55, p. 293–305, doi: 10.1190/1.1442837.

Steinberger, B., Sutherland, R., and O’Connell, J., 2004, Pre-diction of Emperor-Hawaii seamount locations from a revised model of global plate motion and mantle fl ow: Nature, v. 430, p. 167–173, doi: 10.1038/nature02660.

Stewart, J., and Watts, A.B., 1997, Gravity anomalies and spatial variations of flexural rigidity at mountain ranges: Journal of Geophysical Research, v. 102, p. 5327–5352, doi: 10.1029/96JB03664.

Storey, B.C., Leat, P.T., Weaver, S.D., Pankhurst, R.J., Bradshaw, J.D., and Kelley, S., 1999, Mantle plumes and Antarctica-New Zealand rifting; evidence from

mid-Cretaceous mafi c dykes: Journal of the Geo-logical Society, v. 156, p. 659–671, doi: 10.1144/gsjgs.156.4.0659.

Studinger, M., and Bell, R., 2007, Moho topography of the West Antarctic Rift System from inversion of aero-gravity data: Ramifi cations for geothermal heat fl ux and ice streaming, in Cooper, A.K., et al., eds., Ant-arctica: A Keystone in a Changing World—Online Proceedings of the 10th ISAES X, USGS Open-File Report 2007–1047, extended abstract 031.

Studinger, M., Bell, R., Finn, C.A., and Blankenship, D.D., 2002, Mesozoic and Cenozoic extensional tectonics of the West Antarctic Rift System from high-resolution airborne geophysical mapping: Royal Society of New Zealand Bulletin, v. 35, p. 563–569.

Swain, C.J., 1996, Horizontal acceleration corrections in air-borne gravimetry: Geophysics, v. 61, p. 273–276, doi: 10.1190/1.1443948.

Tessema, A., and Antoine, L.A.G., 2003, Variation in ef-fective elastic plate thickness of the East Africa litho-sphere: Journal of Geophysical Research, v. 108, 16 p., doi:10.1029/2002JB002200.

Thomas, R., Rignot, E.J., Casassa, G., Kanagaratnam, P., Acuña, C., Akins, T., Brecher, H., Frederick, E., Gogineni, P., Krabill, W., Manizade, S., Ramamoorthy , H., Rivera, A., Russell, R., Sonntag, J., Swift, R., Yungel , J., and Zwally, J., 2004, Accelerated Sea-Level Rise from West Antarctica: Science, v. 306, p. 255–258, doi: 10.1126/science.1099650.

Trey, H., Cooper, A.K., Pellis, G., Della Vedova, B., Cochrane , G., Brancolini, G., and Makris, J., 1999, Transect across the West Antarctic rift system in the Ross Sea, Antarctica: Tectonophysics, v. 301, p. 61–74, doi: 10.1016/S0040-1951(98)00155-3.

Valliant, H.D., 1992, LaCoste & Romberg Air/Sea Meters: An Overview, CRC Handbook of Geophysical Explo-ration at Sea: London, CRC Press.

Vaughan, D.G., Corr, H., Ferraccioli, F., Frearson, N., O’Hare, A., Mach, D., and Holt, J.W., 2006, New boundary conditions for the West Antarctic Ice Sheet: Subglacial topography beneath Pine Island Glacier: Geophysical Research Letters, v. 33, 4 p., doi:10.1029/2005GL025588.

Vogel, S.W., and Tulaczyk, S., 2006, Ice-dynamical con-straints on the existence and impact of subglacial volcanism on West Antarctic ice sheet stability: Geo-physical Research Letters, v. 33, 4 p., doi: 23510.21029/22006GL027345.

von Frese, R.R.B., Hinze, W.J., Braile, L.W., and Luca, A.J., 1981, Spherical earth gravity and magnetic anomaly modeling by Gauss- Legendre quadrature integration: Journal of Geophysics, v. 49, p. 234–242.

Wannamaker, P.E., Stodt, J.A., and Olsen, S.L., 1996, Dormant state of rifting below the Byrd Subglacial Basin, West Antarctica, implied by magnetotelluric (MT) profi ling: Geophysical Research Letters, v. 23, p. 2983–2986, doi: 10.1029/96GL02887.

Watts, A.B., 2001, Isostasy and fl exure of the lithosphere: Cambridge, Cambridge University Press, 458 p.

Weaver, S.D., Storey, B.C., Pankhurst, R.J., Mukasa, S.B., DiVenere , V.J., and Bradshaw, J.D., 1994, Antarctica-New Zealand rifting and Marie Byrd Land lithospheric magmatism linked to ridge subduction and mantle plume activity: Geology, v. 22, p. 811–814, doi: 10.1130/0091-7613(1994)022<0811:ANZRAM>2.3.CO;2.

Wessel, P., and Smith, W.H.F., 1991, Free software helps map and display data: Eos, Transactions, Ameri-can Geophysical Union, v. 72, p. 441, doi: 10.1029/90EO00319.

Winberry, J.P., and Anandakrishnan, S., 2003, Seismicity and neotectonics of West Antarctica: Geophysical Re-search Letters, v. 30, 3 p., doi:10.1029/2003GL018001.

Winberry, J.P., and Anandakrishnan, S., 2004, Crustal structure of the West Antarctic rift system and Marie Byrd Land hotspot: Geology, v. 32, p. 977–980, doi: 10.1130/G20768.1.

Woollard, G.P., 1979, The new gravity system—Changes in international gravity base values and anomaly val-ues: Geophysics, v. 44, p. 1352–1366, doi: 10.1190/1.1441012.

MANUSCRIPT RECEIVED 18 FEBRUARY 2008REVISED MANUSCRIPT RECEIVED 22 OCTOBER 2008MANUSCRIPT ACCEPTED 23 APRIL 2009

Printed in the USA